中文摘要：

针对红外与可见光成像传感器的物理特性,提出了一种基于非采样Contourlet变换的红外与可见光图像融合算法。首先对原始图像分别进行非采样Contourlet变换,得到不同尺度与方向下的子带系数。对低频子带系数,采用加权平均的融合规则;对不同尺度与方向下的高频子带系数,采用基于局部区域能量匹配的融合规则。最后经非采样Contourlet逆变换得到融合结果。实验结果表明,该算法可以有效地综合可见光与红外图像中的重要信息,其融合结果较典型的基于塔式分解与基于小波变换的图像融合算法,在主观视觉效果与客观评价指标上均有所改善。

英文摘要：

An infrared and visible iamge fusion algorithm based on non-sampling Contourlet transform（NSCT） is put forward according to the physical charcteristics of infrared and visible imaging sensor.The NSCT for the original images is conducted to get the sub-band coefficients at different scales and directions.The weighted average fusion rule is adopted for the low frequency sub-band coefficients,while the fusion rule based on the local area energy matching is employed for the high frequency sub-band coefficient at different scales and directions.Then the image fusion is achieved by the aid of the non-sampling Contourlet inverse transformation.The fusion result shows that the algorithm can effectively composite the important information in visible and infrared images.The fusion result is beter than the typical image fusion algorithm based on the pyramidal decomposition and wavelet transform in subjective vision effect and objective evaluation index.